Submitted by Kimbar on Mon, 2004-03-08 22:04
I've got some idea. I put it here to discuss it with you.
I think that it would be nice if we could write scripts defining new
genotypes formulas. Only few functions will be needed, like: converf0(),
mutate(), crossover(), and so on. To make it easier maybe codification of f1
genotype in script form should be avalitable. I've got some idea of genotype
formula, but to use it I have to change the code of whole application. I
hope that there are no difficulties in puting it into framstick-script form
in separate file.
What do you think about it?
Regards Grzegorz Kimbar
Forums:
> I think that it would be
> I think that it would be nice if we could write scripts defining new
> genotypes formulas. Only few functions will be needed, like: converf0(),
> mutate(), crossover(), and so on. To make it easier maybe codification of f1
> genotype in script form should be avalitable. I've got some idea of genotype
> formula, but to use it I have to change the code of whole application. I
> hope that there are no difficulties in puting it into framstick-script form
> in separate file.
>
> What do you think about it?
Hmm, the f1->f0 conversion (as you can see in the c++ source which is available within the SDK) is not very simple, and the f1 genetic operators are quite sophisticated, so these are not suitable for FramScript. But for simpler genetic encodings, it might be possible... We will see.
MacKo